Graham A McLeod1, Parandoush Abbasian2,3, Darion Toutant4, Amirhossein Ghassemi4, Tyler Duke4, Conrad Rycyk4, Demitre Serletis5,6, Zahra Moussavi4, Marcus C Ng4,7. 1. Department of Clinical Neurosciences, University of Calgary, Calgary, AB, Canada. 2. Medical Physics, Department of Physics and Astronomy, University of Manitoba, Winnipeg, MB, Canada. 3. CancerCare Manitoba Research Institute, Winnipeg, MB, Canada. 4. Biomedical Engineering, University of Manitoba, Winnipeg, MB, Canada. 5. Charles Shor Epilepsy Center, Cleveland Clinic, Cleveland, OH, USA. 6. Department of Neurosurgery, Cleveland Clinic Foundation, Cleveland, OH, USA. 7. Section of Neurology, University of Manitoba, Winnipeg, MB, Canada.
Abstract
STUDY OBJECTIVES: To compare estimated epileptic source localizations from 5 sleep-wake states (SWS): wakefulness (W), rapid eye movement sleep (REM), and non-REM 1-3. METHODS: Electrical source localization (sLORETA) of interictal spikes from different SWS on surface EEG from the epilepsy monitoring unit at spike peak and take-off, with results mapped to individual brain models for 75% of patients. Concordance was defined as source localization voxels shared between 2 and 5 SWS, and discordance as those unique to 1 SWS against 1-4 other SWS. RESULTS: 563 spikes from 16 prospectively recruited focal epilepsy patients across 161 day-nights. SWS exerted significant differences at spike peak but not take-off. Source localization size did not vary between SWS. REM localizations were smaller in multifocal than unifocal patients (28.8% vs. 54.4%, p = .0091). All five SWS contributed about 45% of their localizations to converge onto 17.0 ± 15.5% voxels. Against any one other SWS, REM was least concordant (54.4% vs. 66.9%, p = .0006) and most discordant (39.3% vs. 29.6%, p = .0008). REM also yielded the most unique localizations (20.0% vs. 8.6%, p = .0059). CONCLUSIONS: REM was best suited to identify candidate epileptic sources. sLORETA proposes a model in which an "omni-concordant core" of source localizations shared by all five SWS is surrounded by a "penumbra" of source localizations shared by some but not all SWS. Uniquely, REM spares this core to "move" source voxels from the penumbra to unique cortex not localized by other SWS. This may reflect differential intra-spike propagation in REM, which may account for its reported superior localizing abilities.
STUDY OBJECTIVES: To compare estimated epileptic source localizations from 5 sleep-wake states (SWS): wakefulness (W), rapid eye movement sleep (REM), and non-REM 1-3. METHODS: Electrical source localization (sLORETA) of interictal spikes from different SWS on surface EEG from the epilepsy monitoring unit at spike peak and take-off, with results mapped to individual brain models for 75% of patients. Concordance was defined as source localization voxels shared between 2 and 5 SWS, and discordance as those unique to 1 SWS against 1-4 other SWS. RESULTS: 563 spikes from 16 prospectively recruited focal epilepsy patients across 161 day-nights. SWS exerted significant differences at spike peak but not take-off. Source localization size did not vary between SWS. REM localizations were smaller in multifocal than unifocal patients (28.8% vs. 54.4%, p = .0091). All five SWS contributed about 45% of their localizations to converge onto 17.0 ± 15.5% voxels. Against any one other SWS, REM was least concordant (54.4% vs. 66.9%, p = .0006) and most discordant (39.3% vs. 29.6%, p = .0008). REM also yielded the most unique localizations (20.0% vs. 8.6%, p = .0059). CONCLUSIONS: REM was best suited to identify candidate epileptic sources. sLORETA proposes a model in which an "omni-concordant core" of source localizations shared by all five SWS is surrounded by a "penumbra" of source localizations shared by some but not all SWS. Uniquely, REM spares this core to "move" source voxels from the penumbra to unique cortex not localized by other SWS. This may reflect differential intra-spike propagation in REM, which may account for its reported superior localizing abilities.
Authors: Chris Plummer; Simon J Vogrin; William P Woods; Michael A Murphy; Mark J Cook; David T J Liley Journal: Brain Date: 2019-04-01 Impact factor: 13.501
Authors: Mustafa Aykut Kural; Lene Duez; Vibeke Sejer Hansen; Pål G Larsson; Stefan Rampp; Reinhard Schulz; Hatice Tankisi; Richard Wennberg; Bo M Bibby; Michael Scherg; Sándor Beniczky Journal: Neurology Date: 2020-04-22 Impact factor: 11.800